financial development and textile sector competitiveness ... · munich personal repec archive...

Munich Personal RePEc Archive

Financial Development and Textile

Sector Competitiveness: A Case Study of

Pakistan

Hanif, Muhammad N. and Jafri, Sabina K.

2006

Online at https://mpra.ub.uni-muenchen.de/10271/

MPRA Paper No. 10271, posted 03 Sep 2008 06:26 UTC

Financial Development and Textile Sector

Competitiveness: A Case Study of Pakistan

MUHAMMAD NADIM HANIF

SABINA KHURRAM JAFRI

Kletzer and Bardhan (1987) argue that countries with a relatively well-developed

financial sector have a comparative advantage in industries that rely on external

finance. Beck (2002) and Fanelli and Medhora (2002) find that well-developed

financial sector translates into a comparative advantage in the production of

manufactured goods. There has been no attempt so far to explore the relationship

between the financial development and international trade competitiveness in the

case of Pakistan. We construct Balassa’s Revealed Comparative Advantage (RCA)

index for textile sector of Pakistan. Using ratio of credit extended to the textile

sector to the total non-government credit of the banking system (Textile Credit

Share [TCS]) as proxy for external finance, we estimate long-run relationship

and Error Correction Mechanisms (ECM) between RCA index and TCS while

controlling for other determinants of the international trade competitiveness of

textile sector of Pakistan. In line with the findings of Beck (2002) and Fanelli

and Medhora (2002), our results suggest that recourse to external finance has a

strong positive impact on the country’s textile sector competitiveness both in the

short and the long run, even when we control for traditional determinants of

competitiveness.

( JEL: F14, O16) Keywords: Financial Development, Competitiveness.

Muhammad Nadim Hanif (corresponding author) and Sabina Khurram Jafri are with the

State Bank of Pakistan, I.I. Chundrigar Road, Karachi-74000, Pakistan. Email: nadeem.hanif@sbp.

org.pk and [email protected]

SOUTH ASIA ECONOMIC JOURNAL, 9:1 (2008): 141–158

SAGE Publications Los Angeles � London � New Delhi � Singapore

DOI: 10.1177/139156140700900106

Acknowledgements: The authors thank Riaz Riazuddin, Omar Farooq Saqib, Farooq Arby,

Zulfiqar Hyder, Imran Naveed Khan and Mahmood ul Hassan Khan of State Bank of Pakistan for

their very helpful comments on earlier draft of the article. The authors also thank an anonymous

referee whose comments have been helpful in improving the clarity of the article. Views expressed

in this article are those of the authors and do not necessarily represent those of the institution they

are employed with.

142 / MUHAMMAD NADIM HANIF AND SABINA KHURRAM JAFRI

South Asia Economic Journal 9:1 (2008): 141–158

1. Introduction

Traditional theories of international trade focus on comparative advantage and

factor endowments concepts. Kletzer and Bardhan (1987) highlight the contribu-

tion of some aspects of credit market imperfections to inter-country differences

in patterns of specialization and trade. They show that even when technology

and endowments are identical between countries and there are no economies

of scale, the difference in the level of financial development may lead to higher

external finance premium in one country compared to other countries. This

may provide comparative advantages to those firms which have access to de-

veloped financial markets or institutions. They presumed that more sophisticated

manufactured finished products require more credit to cover selling and distribu-

tion costs than primary or intermediate products. They show that countries

with a relatively well-developed financial sector have comparative advantage in

industries and sectors that rely more on external finance.

Over the last few decades, researchers have shown that financial sector de-

velopment plays significant role in economic growth. Reforming the financial

sector may have implications for the structure of international trade if the level

of financial development is a determinant of a country’s comparative advantage

as Kletzer and Bardhan (1987) highlighted. Thus, the effect of trade reforms on

the structure of international trade might depend on the level of a country’s

financial development. Recent study by Fanelli and Medhora (2002) vindicates

the argument of Kletzer and Bardhan (1987) by finding that comparative ad-

vantage is positively affected by the financial sector development.

There has been no attempt so far to analyze the relationship between the fi-

nancial development and international trade competitiveness for the case of

Pakistan. This article is an attempt to fill this gap by exploring the role of access

to external finance on the country’s textile exports competitiveness.

In Pakistan, focus of trade policy started shifting from import substitution to

export promotion since the seventies. In 1988, Pakistan signed a Structural

Adjustment Programme with the International Monetary Fund (IMF) to address

its balance of payments deficit problems which required an emphasis on greater

liberalization of both imports and exports. The abolition of trade barriers moves

resources to the products in which country has a comparative advantage. While

there are several links between financial development and international trade,

this article attempts to explore the ability of the financial sector to channel savings

to private sector to help overcome liquidity constraints and raise the international

trade competitiveness of the textile sector of Pakistan.1 Our findings suggest

1 The focus of this study is on textile sector, since textile exports capture around 90 per cent

share in the total manufactured exports of the country.

Financial Development and Textile Sector Competitiveness / 143


that recourse to external finance has a strong positive impact on the country’stextile sector competitiveness both in the short and the long run.

The rest of the article is structured as follows. In Section 2, review of literaturerelating to financial development and trade competitiveness is given. Section 3discusses the data series, the model and the methodology utilized. The empiricalresults are presented in Section 4. The last section concludes the article by sum-marizing the findings of this study.

2. Review of Literature

Among the macroeconomic variables that the empirical literature on economicgrowth has identified as being highly correlated with the growth performanceare financial development (Detragiache and Ueda 2004; Khan and Senhadji 2000;Levine 1997) and the degree of openness (for example, Frankel and Romer1999; Sachs and Warner 1995). The development economists have rarely ex-plored the area of relationship between financial development and trade patterns.Kletzer and Bardhan (1987), in their seminal work in this area, show that coun-tries with relatively well-developed financial sector have a comparative advantagein industries that depend on external finance. In the Kletzer and Bardhan (1987)model, even when technology and endowments are identical between the coun-tries and economies of scale are absent, credit market frictions lead to one countryfacing a higher interest rate or rationed credit compared to other countries. Thismay lead to differences in comparative advantages in processed goods which re-quire more working capital, marketing cost or trade finance. Kletzer and Bardhan(1987) presumed that more sophisticated manufactured finished goods requiremore finance to cover selling and distribution costs than primary or intermediategoods.

While exploring the link between financial development and growth, Rajanand Zingales (1998) concluded that in countries with well-developed financialsystems, industries that are naturally heavy users of external finance grow faster.They argue that this result has implications for trade patterns because well-developed financial sector is a source of comparative advantage for a country inindustries that rely more on external finance.

Motivated by the argument of Rajan and Zingales (1998), Beck (2002) ex-plores a link between level of financial development and the level and structureof international trade. He analyzes theoretically a channel through whicheconomy-wide level of external finance determines the commodity structure oftrade balance. The Beck model focuses on the role of finance in mobilizing sav-ings and facilitating large-scale and high return projects. He also finds empiricalevidence supporting his model that a well-developed financial sector translates



into a comparative advantage in the production of manufactured goods. Thus,the level of financial development translates into comparative advantage in in-dustries that are more dependent on external finance.

According to Fanelli and Medhora (2002), the competitiveness of a countrydepends both on the price and non-price factors. For improving the price com-petitiveness, devaluation can prove helpful in the short run. However, the non-price competitiveness can be induced in industries by enhancing the level ofproductivity. They explain that in an environment of efficient financial markets,the financial intermediaries are in a position of impacting the level of innovationby identifying and channelling funds to the most efficient users. The imperfec-tions in the financial market,2 on the other hand, reduce the ability of the financialsector to efficiently channel funds from lenders to the borrowers and that nega-tively impacts the productivity growth. Hence, higher level of financial develop-ment impacts comparative advantage of a country by enhancing the level ofproductivity by identifying entrepreneurs with the best chances of successfullyimplementing innovative production processes.

3. Data, Model and Methodology

In order to measure competitiveness of textile sector of Pakistan, we constructBalassa (1965) index of Revealed Comparative Advantage (RCA).3 This indexmeasures the relative importance of a sector in a country’s total exports withrespect to the relative importance of the same sector in the overall exports in theworld. It is given by the following equation:

=ij i

ij

wj w

X XRCA

X X(1)

where Xij is country i’s exports of sector j, Xi is total exports of country i, Xwj isworld’s exports of sector j and Xw is the overall exports in the world. Thus, theratio in the numerator is the share of sector j’s export in country i’s total exportsand the ratio in denominator is the share of the same sector in world’s overallexports. If RCAij is greater than unity, it means that country i has comparativeadvantage in production in her sector j. The data to construct RCA index istaken from ‘Comtrade’ database of United Nations Statistics Division.

2 In the form of segmentation in the financial market, scarcity of long-term credit for the financing

of private firms, and low capitalization of the stock markets as compared to the size of the economy.3 The time span of this study is 1974–2004 and we have used annual data series.



The evolution of the RCA index is shown in the Figure 1. It shows that

Pakistan’s textile sector exports are inherently competitive in the world market

since the values of the RCA index are greater than unity throughout the period

of analysis. Besides, it also shows that the country achieved continuous

improvement in the level of competitiveness during this period.

Various proxies have been used in the literature to measure financial sector

development in a country. Some studies conducted specifically on finance–trade

relationship have used firm-level data in which the role of financial sector has

been measured by firms’ dependency on external finance. Here we construct a

proxy of financial development by getting an idea from the measures constructed

by King and Levine (1993) while studying finance–growth nexus. They used

four proxies, namely, ratio of the liquid liabilities to GDP, ratio of the assets of

deposit money banks to the total assets of the banking system, ratio of claims on

the non-financial private sector to total domestic credit (excluding credit to money

bank), and the ratio of claims on the non-financial private sector to GDP. Our

focus in this study is on the textile sector of Pakistan and, hence, we can measure

financial sector role in this context as a ratio of the claims on the textile sector to

FIGURE 1

Time Series Plot of RCA Index



the total value added in the textile sector. But we do not have data for the valueaddition of the Pakistan’s textile sector (separately). The other proxy we can thinkof is ratio of total non-government credit to the overall GDP (CGDR). Thiswill not be able to reflect share of the textile sector in the financial developmentprocess.4 Therefore, we use the ratio of credit extended to the textile sector tothe total non-government credit of the banking system5 denoted by Textile CreditShare (TCS). Figure 2 shows the evolution of TCS during the period understudy.

The model relating the role of finance to the textile sector competitivenessmay be of the form:

Log (RCAt) = β0 + β1Log (TCSt) + εt (2)

4 Similar is the problem with the ratio of overall liquid liabilities to GDP (M2GR).5 However, there is one caveat with the use of this proxy: it shows concentration of the credit to

the textile sector rather than the overall development of the financial sector. The best proxy would

be the firm-level data of external finance used by textile sector. But that requires a different approach

for doing this work.

FIGURE 2

Time Series Plot of TCS



where RCA is a measure of competitiveness of textile sector of Pakistan, TCS is

a proxy for finance and εt is the error term. β0 and β1 are intercept and slope

parameters, respectively, in the model. If the estimate of the slope parametersturns out to be positively significant, it means that the greater availability ofexternal finance to textile sector helps a country improve her international tradecompetitiveness in this sector. However, finance is not the only determinant ofa country’s international trade competitiveness in any sector. In order to assessthe impact of finance on the comparative advantage, we have to control forother determinants. The relative strength of a country’s currency vis-à-vis itscompetitors along with its domestic price level is an important source ofcomparative advantage to a country. We can take this fact into account by inclusionof country’s real effective exchange rate (REER6) in the model. It determinesthe impact of country’s exchange rate policy on trade sector. We expect sign ofREER to be negative while we do regression analysis. With the appreciation(depreciation) of real effective exchange rate, our textile sector exports willbecome costly (cheaper) for the international buyers and this will hurt (boost)demand and our exports will shrink (expand) and RCA will decline (rise)subsequently. Furthermore, Pakistan is world’s fourth largest producer of cottonwhich is core input for the textile sector. This natural edge gives the country acomparative advantage over the competitors. To capture this effect, we includecotton production (CTP) as an explanatory variable in the model.7 Thedefinitions and sources of all the variables used in this article are explained inTable A1 of the Appendix. On the basis of the earlier discussion, one may estimatethe following model:

Log (RCAt) = β0 + β1Log (TCSt) + β2Log (REERt)+ β3Log (CTPt) + εt (3)

But, if we examine Table A2, showing matrix of simple correlation coefficientsbetween RCA, TCS, REER and CTP, we observe that there is strong multi-collinearity problem due to higher correlation between REER and CTP (–0.94)than the correlation between RCA and REER (–0.80) as well as RCA and CTP(0.73). We cannot keep any one of the two out of the model as both these vari-ables are significant from the theoretical point of view. REER is the policy variablethat determines the impact of country’s exchange rate policy on its trade sector.On the other hand, cotton produced in Pakistan is the most important input for

6 We defined it in a way that increase in REER means appreciation in our currency as compared

with the trading partners.7 One may argue that variables like wage inflation (WINF), domestic industrial electricity prices

(DIEP) and high speed diesel prices (HSDP) also affect the textile sector competitiveness through

cost of production channel. In this study, we do not find these variables to have statistically significant

impact on textile sector competitiveness in the long run. However, while doing the short-run

analysis, we do find significance of domestic industrial electricity prices (DIEP).



the textile sector and plays an important role in determining country’s trade

pattern and, hence, international trade competitiveness. The solution we suggest

here is to estimate the following two separate models:

Log (RCAt) = β0 + β1Log (TCSt) + β2Log (REERt) + εt (4)

Log (RCAt) = γ0 + γ1Log (TCSt) + γ2Log (CTPt) + εt (5)

β2 in model (4) is expected to be negative as discussed earlier, while γ2 in model

(5) is expected to have positive sign. As mentioned before, if the estimated slope

parameters of TCS turn out to be positively significant, it would indicate that

availability of external finance to textile sector raises the international competi-

tiveness in textile products.

Looking at their graphs (Figures 1 through 4), all of our series RCA, TCS,

REER and CTP are expected to be integrated of order one. Using Augmented

Dickey–Fuller (ADF) approach, we test the stationarity of the time series used

in this analysis. Then we test for the long-run relationships among the variables,

FIGURE 3

Time Series Plot of REER



as postulated in the models (4) and (5), using Engle–Granger two-step proced-

ure.8 Later, from the equilibrium models, we use errors to built Error Correction

Mechanisms (ECM) to see the short-run dynamics of the models. In the ECMs,

we will also test if the variables like wage inflation (WINF), changes in domestic

industrial electricity prices (DIEP) and/or high speed diesel prices (HSDP) have

short-run impact on the textile sector competitiveness through cost of production

channel.

4. Estimated Results

Before the econometric analysis, we do some simple statistical analysis. The

correlation matrix presented in Table A2 of the Appendix shows that for the

case of Pakistan, there is a positive relationship between the textile credit share

in the total non-government credit advanced by the banking sector and the

FIGURE 4

Time Series Plot of CTP

8 We also test for stability of the estimated coefficients, and for normality and homoscedasticity

of the errors term.



RCA index for textile sector of Pakistan; there is a negative relationship between

real effective exchange rate and Pakistan’s textile sector competitiveness; and

cotton production and RCA index for textile sector are positively related.

The first step in the econometric analysis is the univariate time series analysis.

The augmented Dickey–Fuller (ADF) unit root test has been used to test the

time series properties of the data. Intercept and trend have been included on the

basis of their statistical significance in the ADF equation. The results of ADF

test presented in the Table A3 of the Appendix show that all the variables of the

models (4) and (5) are integrated of order one as they are stationary after first

difference.

While we estimate models (4) and (5), we find the problem of autocorrelation

in the estimated equations. To resolve this issue, lagged dependent variable is

included in each of these models. It is important to note that in addition to cap-

turing the dynamic process, lagged dependent variable is also an excellent proxy

for many of the omitted variables.

As seen from the Figure 1, there is some fall in RCA for textile sector in 1992

which reflects a boom year for cotton production all over the world. World

cotton production reached from 87.2 million bales in 1991 to 95.4 million bales

in 1992. Pakistan’s cotton production also increased from the previous level of

9.6 million bales and reached a historic peak of 12.8 million bales in 1992. How-

ever, world share of textile products in overall exports rose more than that of

Pakistan’s textile exports share in the overall exports. Thus, RCA declined in

spite of the fact that cotton production rose by 33.2 per cent against world cotton

production, which rose by 9.4 per cent. To capture this fall in RCA, a dummy

variable for 1992 is included.

Final equations estimated are presented in Table A4 of the Appendix. First

we see whether the variables in each of the two models exhibit long-run relation

or not. As suggested by Engle and Granger (1987), we test for the stationarity of

the errors from estimated models (4) and (5).

The results for both estimated models in (4) and (5) validate our hypothesis

that the credit availability to the textile sector helps improve Pakistan’s

competitiveness in international trade in textile products.9 The elasticity of RCA

with respect to TCS ranges from 0.17 to 0.20. The regression results in Table A4

show that all the variables have expected signs. From the estimated model (4),

we can see that the elasticity of RCA with respect to REER is –0.19. Appreciation

of our currency, as compared with our trading partners, impacts our textile sectors

9 CGDR and M2GR are suitable proxies for financial development in overall economy rather

than in the context of textile sector as we argued in the earlier section. Furthermore, these indicators

of finance were found to be insignificant when we regressed RCA index upon each of these separately

with other determinants of RCA.



competitiveness negatively. This is evident from the statistically significant esti-

mated coefficient of REER in (4). The elasticity of RCA with respect to CTP is

0.10. Further, as expected, the coefficient of CTP from model (5) is found to be

positively significant.10

Post estimation statistical diagnostic tests were applied to the residual series.

These include Jarque-Bera test for normality, LM test for serial correlation,

White tests for heteroscedasticity. Both the models have passed all these tests. In

Figures 5 to 8 we plot cumulative sum and cumulative sum of squares of recursive

FIGURE 5

CUSUM Test for Estimated Model (4)

10 Another variable which we tried to control for is terms of trade (TOT). It is measured by

ratio of unit value of export to unit value of imports. In both the models (4) and (5), we found esti-

mated coefficient of Log(TOT) to be negative (as one should expect) but statistically insignificant

(with p-values 0.14 and 0.20 respectively). Furthermore, the inclusion of Log(TOT) in model (5)

deteriorates both the Akaike Information Criterion (AIC) and the Schwarz Criterion (SC), and

thus we excluded Log(TOT) from the final equations we estimated for this study.



FIGURE 6

CUSUM of Sqs Test for Estimated Model (4)

FIGURE 7

CUSUM Test for Estimated Model (5)



residuals along with their critical lines (at 5 per cent level of significance). These

plots confirm the estimated coefficients’ stability and the stability of the error

variance.

To test for cointegration between the variables in the models, we apply standard

unit root test upon the errors of these models. The results have been presented

in the Table A5. These tests for cointegrating residuals confirm stationary rela-

tionship among the variables as we have postulated in models (4) and (5).

Having established cointegration as a long-run property, ECM is a natural way

of capturing dynamic adjustment to the long run. With the help of ECM we can

also conduct Granger (1969) non-causality tests. We estimate the following

parsimonious models using ERROR 4 and ERROR 5 estimated before, using

models (4) and (5) for which estimated results have been reported in Table A4.

∆Log (RCAt) = φ0 + φ1ERROR 4t–1 + φ2∆Log (TCSt) +

φ3∆Log (NERDt)+ φ4∆Log (DIEPt)+ ηt

(6)

∆Log (RCAt) = ϕ0 + ϕ1ERROR 5t–1 + ϕ2∆Log (TCSt) +

ϕ3∆Log (NERDt)+ ϕ4∆Log (DIEPt)+ νt

(7)

FIGURE 8

CUSUM of Sqs Test for Estimated Model (5)



In models (4) and (5), we have long-run determinants of RCA. In addition tothese determinants, for the short-run analysis using the ECMs, we have also at-tempted to investigate if variables like wage inflation (WINF), changes indomestic industrial electricity prices (DIEP) and/or high speed diesel prices(HSDP) have short-run impact on the textile sector competitiveness throughcost of production channel. The estimated ECM models in (6) and (7) are pre-sented in Tables A6 and A7 respectively. The error terms are significant in boththe models though it is significant in Model (7) at 10 per cent. In both themodels we find that there is a significant positive short-run impact of externalfinance on Pakistan’s competitiveness in international trade in textile products.We do not find statistical support for short-term impact of REER or CTP onRCA index in any of the two estimated ECMs. We, however, found that thechanges in nominal exchange rate11 of US Dollar (NERD) significantly affectchanges in RCA index in the short run. There is also negative significant impactof industrial electricity prices on RCA index in short run. Wage inflation andchanges in high speed diesel prices do not have such impacts (this may be becauseelec-tricity price has already been showing short-run impact).

5. Conclusion

Traditional theories of international trade focus on comparative advantage andfactor endowments. Kletzer and Bardhan (1987) show that even when there areno economies of scale, and technology and endowments are identical betweencountries, a country with relatively developed financial institutions would havecomparative advantages in the production of processed goods requiring moreexternal finance. The recent study by Fanelli and Medhora (2002) vindicatesthe argument of Kletzer and Bardhan (1987) by finding that comparativeadvantage is positively effected by the financial sector development.

In this article we attempt to explore the ability of the financial sector to channelsavings to private sector to help overcome liquidity constraints and raise theinternational trade competitiveness of the textile sector of Pakistan. We constructBalassa’s RCA index for our textile sector. To examine the role of external financeto textile sector in determining its competitiveness, we estimate two models—one based on REER and other on the cotton production (CTP)—along withthe ratio of credit extended to the textile sector to the total non-governmentcredit of the banking system (TCS). We find that there is stable and long-run

relationship between RCA index and TCS in presence of traditional determinants

11 Pak rupees per US Dollar.



of the international trade competitiveness of textile sector of Pakistan. For both

the models we also estimated ECMs. The results suggest that greater access to

external finance has a strong positive impact in the improvement of country’s

textile sector competitiveness, both in the short and the long run, even when

we control for traditional determinants of competitiveness. Thus, the thesis put

forth by Kletzer and Bardhan (1987), that an efficient financial system helps in

improving country’s comparative advantage, is supported by these results for

the case of Pakistan. Competitiveness in the short-run, however, is also impacted

by Pakistan’s bilateral nominal exchange rate with the US and the domestic in-

dustrial electricity prices.

The findings of this article have a strong policy implication as this underlines

the importance of financial sector development for strengthening of comparative

advantage in the largest exporting sector of Pakistan. Therefore, it is impera-

tive for Pakistan to continue with the ongoing financial reforms. In addition,

the implementation of the second generation reforms would further reduce the

cost of doing business in Pakistan and thus lead to improvement in the com-

parative advantages of country’s exports in a highly competitive global external

environment.

Appendix

TABLE A1

Definition and Sources of the Data Used

Variable Description Source

RCA Revealed Comparative Advantage index Authors’ calculation using ‘Comtrade’

database of United Nations Statistics

Division

TCS Share of credit extended to textile sector State Bank of Pakistan

in total non-government advances

extended by the banking sector

REER Real effective exchange rate State Bank of Pakistan

CTP Log of cotton production Handbook of Statistics on Pakistan Economy

by State Bank of Pakistan

WINF Wage inflation in the textile sector Authors’ calculation using textile sector

wages data from 50 Years of Pakistan in

Statistics (Volume IV) by Federal Bureau

of Statistics

HSDP High speed diesel prices 50 Years of Pakistan in Statistics (Volume IV)

DIEP Domestic industrial electricity prices 50 Years of Pakistan in Statistics (Volume IV)

NERD Nominal exchange rate of Dollar State Bank of Pakistan

(Pak Rupees per US$)



TABLE A2

Correlation Matrix

RCA TCS REER CTP

RCA 1.00 0.60 –0.80 0.73

TCS 1.00 –0.31 0.20

REER 1.00 –0.94

CTP 1.00

TABLE A3

Results of Unit Root Analysis

ADF Test Value (p-value∗ in Parentheses) c, t and n Included@

Variable At Levels First Difference At Levels First Diff.

Log(RCA) –2.85 (0.19) –5.33 (0.00) t, c n

Log(TCS) –1.46 (0.82) –4.57 (0.00) t, c n

Log(REER) –2.24 (0.44) –3.75 (0.00) t, c n

Log(CTP) –2.83 (0.20) –7.18 (0.00) t, c n

Log(HSDP) –1.70 (0.73) –4.44 (0.00) t, c c

Log(DIEP) –2.03 (0.56) –5.67 (0.00) t, c c

Log(NERD) –2.89 (0.18) –3.77 (0.00) t, c c

WINF –6.73 (0.00) c

Notes: We used zero lag as suggested by ‘Schwartz Information Criteria’ for lag selection in ADF

test value.∗ MacKinnon’s one-sided p-values.@ c, t and n denotes constant, trend and none (of the two) included in the ADF regression

estimated to test for unit root.

TABLE A4

Results of Regression Analysis—Log (RCA) as Dependent Variable

Model with Log(REER) Model with Log(CTP)

{test-statistic in brackets} {test-statistic in brackets}

(p-value in parentheses) (p-value in parentheses)

Constant 1.61 (0.00) –0.40 (0.05)

Log(TCS) 0.17 (0.00) 0.20 (0.00)

Log(REER) –0.19 (0.00)

Log(CTP) 0.10 (0.00)

Log (RCA(–1)) 0.49 (0.00) 0.56 (0.00)

Dummy 1992 –0.13 (0.02) –0.15 (0.00)

R2 0.92 0.91

F-Stat {72.00}(0.00) {66.93}(0.00)

Jarque-Bera {0.80} (0.67) {1.51} (0.46)

Serial Correlation LM (F) test {1.61}(0.22) {0.54}(0.58)

White (F) test for Heteroscedasticity {1.98}(0.10) {1.81}(0.13)



TABLE A5

Test of Cointegration (Unit Root Test of Error)

DF Test Value (p-value∗ in Parentheses)

At Levels

Error from Model with Log (REER) –4.19 (0.00)

Error from Model with Log (CTP) –4.45 (0.00)

TABLE A6

Results of ECM Model 6

Test-statistic in Brackets and p-value in Parentheses

Lagged Error 4 –5.21 (0.03)

∆Log(TCS) 2.42 (0.05)

∆Log(NERD) 4.12 (0.00)

∆Log(DIEP) –1.28 (0.00)

TABLE A7

Results of ECM for Model 7

Test-statistic in Brackets and p-value in Parentheses

Lagged Error 5 –0.40 (0.10)

∆Log(TCS) 0.25 (0.06)

∆Log(NERD) 0.35 (0.02)

∆Log(DIEP) –0.12 (0.01)

References

BALASSA, BELLA (1965), ‘Trade Liberalization and “Revealed Comparative Advantage”’, Manchester

School of Economics and Social Studies, 33(2): 99–123.

BECK, THORSTEN (2002), ‘Financial Development and International Trade: Is There a Link?’, Journal

of International Economics, 57(1): 107–31.

DETRAGIACHE, ENRICA and KENICHI UEDA (2004), ‘Does Financial Sector Development Help Eco-

nomic Growth and Welfare’, World Economic Outlook, April, Washington.

DICKEY, D.A. and W.A. FULLER (1979), ‘Distribution of the Estimators for Autoregressive Time

Series with a Unit Root’, Journal of the American Statistical Association, 74(366, Part I): 427–31.

ENGEL, ROBERT and CLIVE GRANGER (1987), ‘Co-Integration and Error Correction: Representation,

Estimation and Testing’, Econometrica, 55(2): 251–76.

FANELLI, J.M. and R. MEDHORA (2002), ‘Finance and Competitiveness: Framework and Synthesis’,

in J.M. Fanelli and R. Medhora (eds), Finance and Competitiveness in Developing Countries, pp. 1–19.

Ottawa: International Development Research Centre.

FRANKEL A. JAFFERY and DAVID ROMER (1999), ‘Does Trade Cause Growth?’, American Economic Re-

view, 89(3): 379–99.

GOVERNMENT OF PAKISTAN (1998), 50 Years of Pakistan in Statistics (Volume IV). Islamabad: Federal

Bureau of Statistics.



GRANGER, C. (1969), ‘Investigating Causal Relations by Econometric Models and Cross Spectral

Methods’, Econometrica, 37(3): 424–38.

KHAN, MOHSIN S. and ABDELHAK SENHADJI (2000), ‘Financial Development and Economic Growth’.

IMF Working Paper WP/00/209, December, Washington D.C.

KING G. ROBERT and ROSS LEVINE (1993), ‘Finance and Growth: Schumpeter Might be Right’,

Quarterly Journal of Economics, 108(3): 717–37.

KLETZER, KEN and PRANAB BARDHAN (1987), ‘Credit Markets and International Trade’, Journal of

Development Economics, 27(1–2): 57–70.

LEVINE, R. (1997), ‘Financial Development and Economic Growth: Views and Agenda’, Journal of

Economic Literature, 35(2): 688–726.

RAJAN R. and LUIGI ZINGALES (1998), ‘Financial Dependence and Growth’, The American Economic

Review, 88(3): 559–86.

SACHS, J.D. and A. WARNER (1995), ‘Economic Reform and the Process of Global Integration’,

Brookings Papers on Economic Activity, 1(1): 1–118.

STATE BANK OF PAKISTAN (2005), Handbook of Statistics on Pakistan Economy. Karachi: State Bank of

Pakistan.

UN COMTRADE. http://comtrade.un.org/db/ (accessed in June 2006).

financial development and textile sector competitiveness ... · munich personal repec archive...

Documents